In Vitro Activity of Fenticonazole against Candida and Bacterial Vaginitis Isolates Determined by Mono- or Dual-Species Testing Assays

We determined the in vitro activity of fenticonazole against 318 vaginitis isolates of Candida and bacterial species and selected 28 isolates for time-kill studies. At concentrations equal to 4× MIC, fenticonazole reached the 99.9% killing endpoint by ∼10 h for Staphylococcus aureus, Streptococcus agalactiae, and Escherichia coli and by ∼17 h for Candida albicans and Candida parapsilosis; and at concentrations equal to 8× MIC, by ∼19 and ∼20 h for Candida glabrata and Candida tropicalis, respectively.

ABSTRACT

We determined the in vitro activity of fenticonazole against 318 vaginitis isolates of Candida and bacterial species and selected 28 isolates for time-kill studies. At concentrations equal to 4× MIC, fenticonazole reached the 99.9% killing endpoint by ∼10 h for Staphylococcus aureus, Streptococcus agalactiae, and Escherichia coli and by ∼17 h for Candida albicans and Candida parapsilosis; and at concentrations equal to 8× MIC, by ∼19 and ∼20 h for Candida glabrata and Candida tropicalis, respectively. At concentrations equal to 2× MIC, fenticonazole required ∼20 h to reach the above endpoint against C. albicans in mixed culture with S. aureus, S. agalactiae, or E. coli versus ∼17 h against C. albicans in pure culture. Supra-MICs are achievable in topically treated patients’ vaginal surfaces.

TEXT

Candida vulvovaginitis (here referred to as vaginitis) is a common condition affecting up to 75% of women at least once in their lifetime (1). Candida albicans is the most frequent cause of this nonfatal, yet distressing disease (2), although species such as Candida glabrata, Candida parapsilosis, and Candida tropicalis have recently been associated with vaginitis (3). Notably, azole-resistant Candida isolates may be responsible for recurring and refractory vaginal infections (4), which are debilitating conditions (5). Multiple factors are associated with symptomatic Candida infection, with symptoms likely determined by the yeast’s overgrowth and penetration of vulvovaginal epithelial cells (1). While topical and oral antifungal azoles are equally effective (6), treatment of symptomatic Candida vaginitis with topical azoles may necessitate longer courses to eradicate the infection (7).

Despite occurring rarely (<5%), mixed vaginitis is due to the simultaneous presence of at least two vaginal pathogens (e.g., a bacterium and a Candida species) that require the use of dual (e.g., antibacterial and antifungal) therapy or possibly broad-spectrum monotherapy for complete eradication of concurrent symptoms (8). Although underestimated until recently (9), coinfecting Gram-positive and Gram-negative bacteria are important causes of aerobic vaginitis (10), with Streptococcus species (e.g., Streptococcus agalactiae), Staphylococcus aureus, and Escherichia coli being the most prevalent pathogens identified in symptomatic women (11). Note that S. aureus is strongly attached to the hyphae of C. albicans, so it is less susceptible to antibacterial treatment (12). Accordingly, interaction of C. albicans with S. aureus (or various other bacterial species) may influence the antimicrobial susceptibility of both organisms when evaluated in in vitro assays. This should mimic the results of mixed cultures of bacterial species exposed to antibiotics (13, 14).

Among the available azoles for clinical use (see ref. 15 for historical review), fenticonazole is an antifungal imidazole derivative, which was synthesized and developed by Recordati S.p.A. (Milan, Italy) about 3 decades ago for the topical treatment of dermatomycoses and Candida vaginitis (16). Interestingly, whereas the gamut of activity of fenticonazole comprises bacteria commonly associated with symptomatic vaginal discharge, the intravaginal administration of fenticonazole has shown efficacy in patients with Candida vaginitis, bacterial vaginosis, or mixed infection (17). Therefore, fenticonazole may be a topical alternative to multiagent antimicrobial treatment of mixed bacterial and fungal infections (17).

The previous evaluation of fenticonazole MICs for single cultures of C. albicans and C. glabrata did not include time-kill assays using single or mixed (i.e., with bacteria) cultures of Candida species (18). In the present study, we first investigated the dynamics of fenticonazole-induced killing in monospecies testing assays for vaginal Candida and bacterial isolates. Then, we determined the fenticonazole killing activity for dual species using mixed cultures of C. albicans with S. aureus, E. coli, or S. agalactiae.

We obtained an anonymized series of microbial isolates from symptomatic women with vaginitis throughout 18 months from the clinical microbiology laboratory at the Fondazione Policlinico Universitario A. Gemelli IRCCS of Rome (Italy). At the collection time, the isolates were identified using matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) and stored at −80°C until testing. We serially diluted fenticonazole [1-(2,4-dichlorophenyl)-2-(N-imidazolyl) ethyl 4-phenylthiobenzyl ether nitrate; Recordati] from a 100× stock solution in dimethyl sulfoxide (Merck KGaA, Darmstadt, Germany) to achieve a 0.008- to 128-μg/ml final concentration range. Antimicrobial susceptibility testing was performed according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) documents for testing yeasts (http://www.eucast.org/ast_of_fungi/) and bacteria (http://www.eucast.org/ast_of_bacteria/). Antifungal MICs were the minimum concentration that yielded ≥50% growth reduction (http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/AFST/Antifungal_breakpoints_v_7.0.pdf). The testing media were either RPMI 1640 plus 2% glucose broth buffered with morpholinepropanesulfonic acid or unsupplemented cation-adjusted Mueller-Hinton (MH) broth. The control strains were C. parapsilosis ATCC 22019, Candida krusei ATCC 6258, S. aureus ATCC 25923, and E. coli ATCC 25922. The time-kill experiments were determined using inocula from exponential growth cultures obtained for each isolate starting from an overnight culture at 37°C in agitation. For pure-culture time-kill experiments, we used 100 μl of the exponentially growing cultures in 5 ml of prewarmed RPMI 1640 or MH medium with or without (control) an antifungal to reach an inoculum of 1 × 10⁵ to 5 × 10⁵ CFU/ml. For mixed-culture time-kill experiments, we inoculated 100 μl of each of two desired microbial species (1:1) into the same MH medium volume. The fenticonazole concentrations tested were equivalent to 0.5×, 1×, 2×, 4×, and 8× the MIC value, as determined for each species. Control and each test solution vials were incubated at 37°C with agitation in an Eppendorf ThermoMixer C (Eppendorf AG, Hamburg, Germany). A total of 100-μl aliquots from control and test solution vials for microbial quantification were removed at 0, 2, 4, 6, 8, 12, and 24 h. The CFU per milliliter were obtained as 10-fold serial dilutions in phosphate-buffered saline and plated (50-μl samples) onto Trypticase-soy (bacterial species) or Sabouraud (fungal species) agars. The plates were incubated at 37°C for 24 to 48 h. The lower limit of accurate and reproducible microbial quantitation was 20 CFU/ml. Each experiment was performed three times to increase statistical significance and evaluated the antifungal carryover effect using the method reported by Cantón et al. (19). According to previous studies (19–21), we analyzed and compared killing kinetics by fitting time-kill data to the exponential equation N_t = N_o × e^−Kt (t, incubation time; N_t, viable cells at time t; N_o, starting inoculum; K, killing rate). The exponential equation transformed into a line by applying natural logarithms (log N_t = log N_o + Kt) assessed the goodness of fit of the data by the r² value (>0.8). The fungicidal activities were compared by the K values that indicated killing or growth. Thus, the K values required to calculate time in hours to reach 50% (t₅₀, 0.30103/K), 90% (t₉₀, 1/K), 99% (t₉₉, 2/K), and 99.9% (t_99.9, 3/K) reductions in growth were compared with those for the starting inoculum.

Table 1 shows the fenticonazole MIC results for the 318 vaginal isolates tested. MICs for S. aureus (n = 35) were slightly higher than those for the other species, including E. coli (n = 25), S. agalactiae (n = 30), C. albicans and other Candida species (n = 186, in total), and Gardnerella vaginalis (n = 42). Conversely, S. agalactiae, C. parapsilosis, and G. vaginalis were the most susceptible tested species (MIC₉₀, 0.03 μg/ml). As for Candida species, C. albicans (n = 51; MIC₉₀, 0.06 μg/ml) and C. parapsilosis (n = 52; MIC₉₀, 0.03 μg/ml) isolates had slightly lower MICs than C. glabrata (n = 44; MIC₉₀, 0.25 μg/ml) and C. tropicalis (n = 39; MIC₉₀, 0.125 μg/ml) isolates. For use in the monospecies killing assays, we randomly selected 28 isolates of C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, S. aureus, S. agalactiae, and E. coli (4 isolates/species). Fenticonazole MICs (μg/ml) were 0.125 (2 isolates) and 0.25 (2 isolates) for C. albicans, 0.125 (1 isolate) and 0.25 (3 isolates) for C. glabrata, 0.03 (2 isolates) and 0.06 (2 isolates) for C. parapsilosis, 0.06 (1 isolate) and 0.125 (3 isolates) for C. tropicalis, 0.25 (2 isolates) and 0.5 (2 isolates) for S. aureus, 0.06 (2 isolates) and 0.125 (2 isolates) for S. agalactiae, and 0.06 (1 isolate), 0.125 (2 isolates), and 0.25 (1 isolate) for E. coli.

TABLE 1.

Susceptibility of 318 vaginal isolates to fenticonazole determined by the EUCAST broth microdilution method

Speciesa (no. of isolates)	MIC (μg/ml)
	MIC90	MIC50	Range
E. coli (25)	0.125	0.03	0.016–0.25
S. agalactiae (30)	0.03	0.016	0.008–0.125
S. aureus (35)	1	0.5	0.25–4
C. albicans (51)	0.06	0.03	0.016–0.25
C. glabrata (44)	0.25	0.125	0.06–0.5
C. tropicalis (39)	0.125	0.06	0.03–0.25
C. parapsilosis (52)	0.03	0.016	0.008–0.125
G. vaginalis (42)	0.03	0.016	0.008–0.03

^aMueller-Hinton and RPMI 1640 broths were the media used for performing antimicrobial susceptibility testing of the listed species of bacteria and Candida, respectively, according to EUCAST recommendations.

Figure 1 depicts the time-kill curves of each of the seven species (means and standard deviations) tested as pure cultures, and Table 2 provides time-killing results for the same Candida and bacterial species. At fenticonazole concentrations of equal to 4× MIC for all species, the number of CFU/ml decreased >3 log units (99.9% killing) by incubation of 16.95 and 17.30 h for C. albicans and C. parapsilosis, respectively, and of 10.60, 10.30, and 10.31 h for S. aureus, S. agalactiae, and E. coli, respectively (Table 2). Compared to results in C. albicans and C. parapsilosis isolates, fenticonazole achieved a ≥3-log decrease (killing endpoint) with a concentration of equal to 8× MIC by 18.59 h for C. glabrata and 20.41 h for C. tropicalis (Table 2) isolates. Interestingly, the lowest fenticonazole concentration to achieve the killing endpoint was 2× MIC at 17.34 h only for C. albicans isolates (Table 2).

FIG 1.

Fenticonazole time killing of Candida (C. albicans, C. glabrata, C. parapsilosis, and C. tropicalis) and bacterial (S. aureus, S. agalactiae, and E. coli) species in pure cultures. Error bars, means ± SD of independently determined values for four isolates of each indicated species. Broken lines, ≥99.9% growth reduction compared with that of the initial inoculum (fungicidal effect).

TABLE 2.

Mean times to achieve 50%, 90%, 99%, and 99.9% growth reductions from the initial inoculum at the indicated multiple of the MIC of fenticonazole

Isolate tested, by culture type	Mean time to achieve growth reduction (h) ofa :
	t50			t90			t99			t99.9
	2× MIC	4× MIC	8× MIC	2× MIC	4× MIC	8× MIC	2× MIC	4× MIC	8× MIC	2× MIC	4× MIC	8× MIC
Pure culture
C. albicans	1.74	1.70	1.93	5.78	5.65	6.43	11.56	11.31	12.85	17.34	16.95	19.28
C. glabrata	NAb	>48	1.87	NA	>48	6.20	NA	>48	12.39	NA	>48	18.59
C. parapsilosis	10.45	1.74	1.87	34.72	5.77	6.21	>48	11.53	12.41	>48	17.30	18.62
C. tropicalis	NA	NA	2.05	NA	NA	6.80	NA	NA	13.61	NA	NA	20.41
S. aureus	1.21	1.06	0.94	4.03	3.53	3.14	8.06	7.06	6.27	12.10	10.60	9.41
S. agalactiae	1.12	1.03	0.92	3.73	3.43	3.04	7.46	6.87	6.09	11.19	10.30	9.13
E. coli	1.12	1.03	0.94	3.72	3.44	3.13	7.43	6.87	6.26	11.15	10.31	9.40
Mixed culture
C. albicans	2.08	1.57	1.86	6.93	5.22	6.18	13.87	10.45	12.36	20.81	15.68	18.54
+ S. aureus	1.16	1.02	0.86	3.86	3.40	2.87	7.71	6.80	5.74	11.57	10.19	8.62
C. albicans	2.03	1.87	1.77	6.74	6.20	5.89	13.48	12.39	11.79	20.22	18.58	17.68
+ S. agalactiae	−c	1.07	0.99	−	3.54	2.99	−	7.08	6.57	−	10.62	9.86
C. albicans	2.05	2.00	1.78	6.83	6.64	5.91	13.66	13.28	11.83	20.49	19.93	17.74
+ E. coli	1.16	1.02	0.86	3.86	3.40	2.87	7.71	6.80	5.74	11.57	10.19	8.62

^at₅₀, t₉₀, t₉₉, and t_99.9, time to achieve 50%, 90%, 99%, and 99.9% growth reduction, respectively. MICs for isolates tested as pure cultures ranged from 0.03 to 0.25 μg/ml for Candida isolates and 0.06 to 0.25 μg/ml for bacterial isolates. The drug concentrations used in the mixed cultures were equivalent to multiples of the MIC obtained with the C. albicans pure culture (0.25 μg/ml). The MIC value was, respectively, equal to or higher than those of the S. aureus (MIC, 0.25 μg/ml), E. coli (MIC, 0.25 μg/ml), and S. agalactiae (MIC, 0.125 μg/ml) isolates tested together with the C. albicans isolate in separate mixed cultures.

^bNA, not achieved.

^c−, S. agalactiae isolate (MIC, 0.125 μg/ml) was not tested at 2× MIC (i.e., 0.5 μg/ml), which would correspond to an MIC value of 0.25 μg/ml for that isolate.

To determine interspecies effects in the presence of fenticonazole, time-kill curves were obtained by cocultivating C. albicans (MIC, 0.25 μg/ml) with S. aureus (MIC, 0.25 μg/ml), S. agalactiae (MIC, 0.125 μg/ml), or E. coli (MIC, 0.25 μg/ml) isolates. Before assessment of dual-species killing, we assessed that either the MIC or the growth of C. albicans isolates in the MH broth was the same as in the RPMI broth. As shown in Fig. 2, the fungicidal effect of fenticonazole in the three mixed cultures was similar to that for the pure cultures of C. albicans isolates (see Fig. 1). However, fenticonazole reached the fungicidal endpoint (99.9% killing) faster at 4× MIC in mixed cultures of C. albicans and S. aureus isolates than in pure cultures (15.68 versus 16.95 h, respectively) (Table 2). At concentrations of 2× MIC, fenticonazole required ∼20 h to reach the fungicidal endpoint against mixed cultures of C. albicans and S. aureus (20.81 h), S. agalactiae (20.22 h), or E. coli (20.49 h) isolates. These findings suggest that those bacterial species may interfere with the activity of fenticonazole against C. albicans infection. Conversely, at 4× MIC, fenticonazole reached the 99.9% bacterial killing endpoint at 10.19, 10.62, and 10.19 h for S. aureus, S. agalactiae, and E. coli isolates, respectively (Table 2). Because these values were the same as those for bacterial pure cultures, our data suggest that the presence of C. albicans isolates may hinder the bacterial protection by fenticonazole.

FIG 2.

Fenticonazole time killing of C. albicans plus S. aureus (A), C. albicans plus S. agalactiae (B), and C. albicans plus E. coli (C) isolates in mixed cultures. The curves represent fungal (not bacterial) killing, and, accordingly, indicated MICs are those of C. albicans. Error bars, means ± SD of independently determined values for two isolates (one plus one) of each indicated species pair. Broken lines, ≥99.9% growth reduction compared with that of the initial inoculum (fungicidal effect).

Figure 3 shows the effects of fenticonazole concentrations on the killing rates (K values) for C. albicans, C. glabrata, C. parapsilosis, and C. tropicalis infection alone or when the mixture included C. albicans, S. aureus, S. agalactiae, or E. coli isolates. Fenticonazole showed the maximum K value against C. albicans (−0.17 CFU/ml/h) at 2× MIC, whereas we observed the highest values (−0.14, −0.16, and −0.14 CFU/ml/h, respectively) at 8× MIC for C. glabrata, C. parapsilosis, and C. tropicalis isolates. The maximum K value for C. albicans plus S. aureus (−0.19 CFU/ml/h), C. albicans plus S. agalactiae (−0.16 CFU/ml/h), or C. albicans plus E. coli (−0.15 CFU/ml/h, respectively) was achieved at 4× MIC for the fungal species.

FIG 3.

Fenticonazole killing rates (K) for pure cultures of C. albicans, S. aureus, S. agalactiae, and E. coli (A) and for mixed cultures of C. albicans with S. aureus, S. agalactiae, or E. coli (B). In panel B, curves represent fungal (not bacterial) killing, and, accordingly, indicated MICs are those of C. albicans. Error bars, means ± SD of independently determined values for the isolates of each indicated species. Values above the broken lines indicate growth, and values below the broken lines indicate killing.

In 2009, Antonopoulou et al. (18) found excellent agreement between EUCAST and CLSI methods when testing fenticonazole against Candida isolates (249 C. albicans and 11 C. glabrata) from vaginal infections. Similar to our data, EUCAST MIC₅₀ values ranged from 0.03 to 0.25 μg/ml for C. albicans isolates and from 0.03 to 0.5 μg/ml for C. glabrata isolates (18). Fenticonazole inhibits the ERG11 product (the key enzyme in the biosynthetic pathway of ergosterol), leading to fungal cell membrane alterations; it also impairs the release of secreted aspartyl proteinase in C. albicans isolates and blocks the fungal cytochrome oxidases and peroxidases (15, 17). Additionally, fenticonazole completely inhibits C. albicans filamentation (i.e., hyphal formation) at the MIC or higher drug concentrations (15). As for antibacterial activity, we confirmed that the fenticonazole activity reported by Jones et al. (22) against organisms associated with bacterial vaginal (including the prototype G. vaginalis) and skin infection (i.e., S. aureus and streptococci). In that study, MIC₅₀/MIC₉₀ values were 1/2 μg/ml for S. aureus (20 isolates) and ≤0.03/0.125 μg/ml for streptococci (24 isolates), and the MIC for 6 isolates was 0.125 μg/ml for group B streptococci, including S. agalactiae. The antibacterial activity of fenticonazole relates to the selective production of a cytotoxic oxidative metabolite (17), which is similar to the induction of reactive oxygen species (ROS) in C. albicans by miconazole (23), which is also widely used for vulvovaginal candidiasis (7).

Aerobic vaginitis and candidiasis are common pathological conditions associated with vaginal discharge, ranking second and third after bacterial vaginosis. The empirical treatment of symptomatic women requires the use of an agent that has combined antibacterial and antifungal activities (16). In this context, comparative clinical studies showed that fenticonazole is more for topical than for other treatments. The azoles are usually fungistatic, but under certain conditions and formulations, some azoles, such as miconazole, can be fungicidal (23, 24). Like amphotericin and ciclopirox (different drug classes), miconazole induces a common oxidative-damage cellular death pathway that culminates in the formation of lethal ROS (25). ROS induction also underlies fenticonazole microbicidal activity as shown in our study, but it requires future investigation.

In conclusion, our MIC data reinforce the potent in vitro activity of fenticonazole against Candida and bacterial species, whereas time-kill data highlight its potential microbicidal at supra-MICs easily reached in topically treated patients for skin/mucosal surfaces. Additionally, we show that increasing fenticonazole concentrations allowed the potential interference between C. albicans and S. aureus or the other bacterial species evaluated as mixed cultures to be overcome. Finally, we recommend the use of topical fenticonazole for the treatment of vaginal infections with different etiology, which may minimize the risk of selecting drug-resistant microbial strains.